Egyptian Journal of Petroleum xxx (xxxx) xxx

Contents lists available at ScienceDirect

Egyptian Journal of Petroleum

journal homepage: www.sciencedirect.com

Full Length Article

Optimization of natural gas treatment for the removal of CO2 and H2S in
a novel alkaline-DEA hybrid scrubber q
Samuel Eshorame Sanni a,⇑, Oluranti Agboola a, Omololu Fagbiele a, Esther Ojima Yusuf a,
Moses Eterigho Emetere b
a
Department of Chemical Engineering, Covenant University, P.M.B 1023, Ota, Ogun State, Nigeria
b
Department of Physics, Covenant University, P.M.B 1023, Ota, Ogun State, Nigeria

a r t i c l e i n f o a b s t r a c t

Article history: Contaminated natural gas when carelessly handled, often poses human and equipment related problems
Received 21 February 2019 ranging from lung and skin infections to corrosion, equipment fouling/failure and reduction in gas quality
Revised 6 November 2019 owing to the presence of acid gases. In this work, four natural gas (NG) samples were treated to remove
Accepted 21 November 2019
CO2 and H2S using 10–50% Di-Ethanolamine (DEA) solutions mixed with 5% w/w 0.1 M calcium hydrox-
Available online xxxx
ide. The treatment process gave increased acid gas removal at increased DEA concentrations. Based on the
simulation results, cost effective treatment of the gas samples, require 0.1 M Ca(OH)2 and DEA mixed
Keywords:
solutions in the range of 27.4–30%. The optimum mixture concentration for the gas treatment was found
Absorption
Acid gas
to be 30% Ca(OH)2-DEA hybrid solution with feed gas flow rate of 830 kscf/h. In terms of pressure energy
Gas treatment consumption, pumping the hybrid mix at 830 kscf/h will save pressure energy as compared to pumping
Process optimization the feed gas at 1024.58 kscf/h since the lower and upper limit feed gas flow rates gave similar results. The
Process simulation optimum pressure for NG treatment was found to be in the range of 2–2.7 bar (2–2.7*105 kgm
1 s
2).
Ó 2019 Egyptian Petroleum Research Institute. Production and hosting by Elsevier B.V. This is an open
access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction
The risks posed by CO2, H2S and other impurities in Natural Gas
(NG) are increasingly alarming owing to their negative conse-
quences in humans, equipment and the environment. Natural gas
may be classified as sweet or sour. It comprises of methane, ethane,
propane, isobutene, n-butane, nitrogen, CO2, O2, isopentane,
n-pentane, hexane and H2 [1]. Other gases include helium, hydro-
gen sulphide and mercaptans which give the gas its characteristic
odour. CO2 and H2S are the major pollutants in NG. For easy/safe
transportation, <50 ppm CO2 is desired [2]. CO2 and H2S can be
trapped using amine solvents, absorption equipment and mem-
branes [3]. Other contaminants, such as carbonyl sulphide, mer-
captans, ethane, pentane etc., are usually removed via distillation
and absorption. Gas purification depends on the target-solute sol-
ubility, partial vapour pressures of the constituents and the spent
heat during solvent recovery [4,5]. Absorption of CO2 from flue
gas using several alcohol-amine solvents has been reported [6].
Peer review under responsibility of Egyptian Petroleum Research Institute.
⇑ Corresponding author.
E-mail addresses: adexz3000@yahoo.com (S. Eshorame Sanni), oluranti.agboo-
la@covenantuniversity.edu.ng (O. Agboola), omololu.oni@covenantuniversity.edu.
ng (O. Fagbiele), esther.yusuf@covenantuniversity.edu.ng (E.O. Yusuf), moses.
emetere@covenantuniversity.edu.ng (M.E. Emetere).
According to Fang and Zhu [7], the use of amines, carbonates, aque-
ous ammonia, polymer membranes, ionic liquids and enzymes are
recent advances in gas treatment operations. Acid gas removal
from contaminated gas depends on the desired application; in
internal combustion engines, a CH4 concentration >90% is friendly
[8,9]. High CO2 in NG reduces engine power [10,11], while
~3500 ppm H2S in gaseous fuels may cause internal corrosion of
engines [12]. High amount of CO2 also reduces the burn-rate of
natural gas [3]. Transportation of significant amounts of H2S can
cause pipeline corrosion [13], leakages, fire explosions and loss of
aquatic and human lives [14] hence, the gas must be treated prior
transportation [15]. In selecting a solvent for gas treatment, one
must consider its solubility, viscosity, solvent corrosivity, density,
thermal stability, % H2S/CO2 in feed gas, process economics and
solvent recovery [16,17]. According to reports from National
Energy Laboratory, the recommended residual sulphur threshold
in NG is 0.1 ppm [18]. One popular amine for acid gas absorption
is Di-ethanolamine (DEA) with formula HN(CH2CH2OH)2 [19,20].
Natural gas contains significant quantities of H2S and CO2 [21]
thus, controlling these contaminants is critical in quantifying the
risks associated with equipment fouling. Natural gas liquefies at

161 °C and 1 atm which necessitates CO2 removal [22,23].
Fig. 1 illustrates a traditional gas treatment process (i.e. the Rec-
tisol process) where cold CH3OH is used to absorb acid gases from

https://doi.org/10.1016/j.ejpe.2019.11.003
1110-0621/Ó 2019 Egyptian Petroleum Research Institute. Production and hosting by Elsevier B.V.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Please cite this article as: S. Eshorame Sanni, O. Agboola, O. Fagbiele et al., Optimization of natural gas treatment for the removal of CO2 and H2S in a novel
alkaline-DEA hybrid scrubber, Egyptian Journal of Petroleum, https://doi.org/10.1016/j.ejpe.2019.11.003
2 S. Eshorame Sanni et al. / Egyptian Journal of Petroleum xxx (xxxx) xxx
Nomenclature
Symbols Designations Units
Letters
F DEA flow rate kgmole/h
G Gas flow rate kscf/h
T Temperature oC
P Pressure kPa, bar
U Heat transfer coefficient kWK
1
A Heat transfer area m2
DP Change in pressure kPa
DT Change in temperature K
Fig. 1. A typical absorption process at atmospheric pressure.
contaminated gas [4,24–26]. Another approach is the Fluor process
which takes advantage of CO2 partial pressure and propylene car-
bonate (C4H6O3) as solvent for removing CO2. The Selexol process
uses dimethyl ethers of polyethylene glycol to trap acid gases at
2.07–13.8 MPa. According to Tatin et al. 90% gas-purification is
possible [27]. Other modelling and simulation works recorded
higher but <98% gas purity [28–35]. In this study, a novel hybrid
DEA and Ca(OH)2 mixture, not previously investigated, was tested
for its ability to remove acid gases in order to meet lower pumping
costs and product specification of <1% vol/vol of H2S and CO2. An
amine package (AMSIM) with adaptable features to the Kent-
Eisenberg model [36–38] was adopted for the simulation.
2. Materials and methods
Equipment and Reagent: Conval Clampseal throttling valve
(USA), PIBV2 post indicator and butterfly target valve supervisory
switch (Ohio, USA), AYVAZ AFBT flash tank (Istanbul UAE), multi-
stream heat exchanger (Linde Engineering, Germany), Gas absor-
ber/Stripper (Rufouz Hitek Engineers Pvt. ltd., Mumbai, India) sep-
arator (UOP7MKII, Armfield), 20-tray regenerator (Haldor Topsoes,
Denmark), DEA- Ca(OH)2 mixer (Transtech Fabrication, Texas USA),
centrifugal pump (EDUR GmbH, Germany), cooler and reboiler
(Industrial Heat Transfer Inc.). 10 mg of 26.6 mL of 99% DEA solu-
tion (Ludwighafen, Germany) and 0.1 M, 5 w/w% Ca(OH)2 of 96%
purity (Muby Chemicals, India).
2.1. Process configuration
Four gas samples from fields W, X, Y and Z as given in Tables 1a
and 1b were separately released from cylinders under pressure and
Please cite this article as: S. Eshorame Sanni, O. Agboola, O. Fagbiele et al., Optimization of natural gas treatment for the removal of CO2 and H2S in a novel
alkaline-DEA hybrid scrubber, Egyptian Journal of Petroleum, https://doi.org/10.1016/j.ejpe.2019.11.003
Subscripts
DEA DEA concentration by volume %
M.f CO2 Mole fraction of CO2 %
M.f H2S Comp Mole fraction of H2S ppm
Abbreviations
NG Natural gas
DEA Di-ethanolamine
MEA Methanolamine
DGA Di-glycolamine
sent to the absorber; the method adopted here uses an alkaline-
amine hybrid mixture with possibility of attaining gas purity
>99%, <0.1% CH4 loss and high solvent recovery [17]. Lean amine
in parts of 10–50% DEA + 0.1 M Ca(OH)2 mix were prepared sepa-
rately in the mixer. Afterwards, each stream was pumped to the
regenerator and allowed to flow through the heat exchanger
(HE1) and cooler respectively. The mixture from the cooler was
then allowed to flow counter-currently with samples W-Z in the
absorber (Fig. 2). However, prior to conducting the experiment
for the hybrid mix, it is necessary to carry out the following control
experiments in order to embrace the necessity of combining both
solvents (DEA and Ca(OH)2) as hybrid mix. For the control experi-
ments, the procedure described above was carried out for gas sam-
ple 4 (i.e. sample D) using DEA only as well as (CaOH)2 only as
separate solvents pumped from the mixer to the regenerator and
then to the absorber. Gas sample 4 was chosen because all gas
samples have near-similar/similar compositions.
After each run, on hourly basis, sweet gas was recovered from
the top of the absorber while rich amine-alkaline mix from the
absorber bottom was allowed to flow through a throttle valve into
a separator where CH4, H2 and trace amounts of CO2 were recov-
ered. The liquid at the bottom of the separator was then contacted
with the incoming regenerator hybrid-mix in HE1. From HE1, the
mix was sent to the regenerator for solvent recovery. Acid gases
were stripped off the rich amine-alkaline mix by raising the regen-
erator temperature from 93.33 to 131.56 °C. The recovered
alkaline-amine was sent back as recycle-stream to the absorber
in order to save energy while ensuring a continuous operation.
Each amine + Ca(OH)2 mix was taken per operation from the sam-
ple point at the regenerator for analysis. Table 1a gives the gas
assays for samples 1–4 (S1-4) from fields W, X, Y and Z
 S. Eshorame Sanni et al. / Egyptian Journal of Petroleum xxx (xxxx) xxx 3

Fig. 2. Flow diagram of the absorption process.

Table 1a
Composition of Natural Gas (Feed Gas) from Fields W, X, Y and Z.

Mole fraction (
)

Components Gas Sample 1 Gas Sample 2 Gas Sample 3 Gas Sample 4
Hydrogen Sulphide 0.13 0.100 0.11 0.13
Carbon dioxide 0.18 0.240 0.22 0.18
Methane 0.69 0.660 0.67 0.69
Ethane 0.000 0.000 0.000 0.000
Propane 0.000 0.000 0.000 0.000
i-Butane 0.000 0.000 0.000 0.000
n-Butane 0.000 0.000 0.000 0.000
i-Pentane 0.000 0.000 0.000 0.000
n-Pentane 0.000 0.000 0.000 0.000

respectively. The %H2S in the samples range from 10 to 13 %; those
of CO2 and CH4 are 18–24% and 66–69% respectively.
2.2. Process simulation
Feed gas compositions as given in Table 1b, were added to the
simulation environment per simulation run. Table 2a gives the con-
ditions in the absorber (plant data) before and after treatment. A
separator with inlet temperature and pressure of 35 °C and
1650 kPa was added with estimated flow rates (Table 3). Standard
ideal liquid volume flow = 43 m3/h, % DEA-hybrid mass fractions
were specified as 10–50% whilst taking into cognisance, prototype
solvent specifications [39]. Number of absorber stages = 20, top
pressure = 3617 kPa, bottom pressure = 3548 kPa with 40 and
70 °C as their corresponding temperatures. Tray weir height, length
and diameter are 0.082, 3.3 and 4 ft respectively. %H2S and CO2 in the
treated gas were estimated (Table 2a). Rich DEA-hybrid leaves the
absorber at 482.63 kPa. Regenerator feed temperature = 93.3 °C.
DP = 70 kPa across HE1. The regenerator consists of 20 stages. A
damping factor of 0.4 was selected to control the step size and allow

Table 1b
NG/Feed Gas-Composition from Fields W, X, Y and Z Adopted for HYSYS Simulation.

Mole fraction (
)

Components Gas Sample 1 Gas Sample 2 Gas Sample 3 Gas Sample 4
Hydrogen Sulphide 0.129 0.095 0.11 0.127
Carbon dioxide 0.178 0.238 0.22 0.175
Methane 0.686 0.660 0.667 0.689
Ethane 0.0031 0.0030 0.0033 0.0030
Propane 0.0022 0.0020 0.002 0.0021
i-Butane 0.00022 0.0002 0.00021 0.0002
n-Butane 0.00019 0.0002 0.00015 0.00022
i-Pentane 0.00004 0.000044 0.00042 0.00041
n-Pentane 0.00004 0.00037 0.00004 0.000039
C7+ 0.00003 0.00004 0.00003 0.00003
Hydrogen Trace Trace Trace Trace
Nitrogen 0.002 0.004 0.004 0.003
Oxygen Trace Trace 0.0001 0.0001

Please cite this article as: S. Eshorame Sanni, O. Agboola, O. Fagbiele et al., Optimization of natural gas treatment for the removal of CO2 and H2S in a novel
alkaline-DEA hybrid scrubber, Egyptian Journal of Petroleum, https://doi.org/10.1016/j.ejpe.2019.11.003
4 S. Eshorame Sanni et al. / Egyptian Journal of Petroleum xxx (xxxx) xxx

Table 2a
Process Stream Conditions for Plant DEA-hybrid Treatment of NG.

Parameter Gas Sample A Gas Sample B Gas Sample C Gas Sample D Gas sample A Gas Sample B Gas sample C Gas sample D
Before Before Before Before After treatment After treatment After treatment After treatment
treatment treatment treatment treatment 10% 20% 30% 35%
M.Flow (kscf/h) 1024.58 2207.13 829.4 1027.3 1024.58 2203.1 830 1027.8
T (⁰C) 81.5 81.5 81.5 81.5 52.8 52.8 52.8 52.8
P (kPa) 6390 5270 7230 6390 5310 5270 5250 6040
Valve Opening 3/5 4/5 2/5 3/5 3/5 4/5 2/5 2/5
CO₂ (mole frac) (-) 0.18 0.24 0.22 0.18 0.07 0.11 0.05 0.05
H₂S (ppm) 58.2 58.8 58.5 58.8 1.08 1.21 1.09 0.86
NO (-) Trace Trace Trace Trace – – – –
CN (-) – – – – – – – –

Table 2b
Process Plant Pre- and Post-natural Gas Treatment with 30% DEA + Ca(OH)2 (Experimental Data).

Plant section 1
Equipment Pressure (kPa) Temperature (°C) No. of Trays
Absorber (Top tray) 3535 43.3
Absorber (Bottom tray) 3549 37.8
Separator 482.6 66.2 20
Amine Regenerator (Top Tray) 241.3 43.3
Amine Regenerator (Bottom Tray) 262 131.6
Plant section 2
DT (°C) DP (kPa) Heat Duty (kW) Adiabatic Eff (%) UA (kWK
1)
Pump 41.2 3390 69.7 75 –
Cooler
63.96 0 228.5 – –
Heat exchanger 0.42 °C 34.47 93.19 – 2.37

Table 3
Equipment Operating Conditions for DEA Hybrid Treatment of Natural gas (Simulation Results).

Stream Temperature (oC) Pressure (kPa) Molar Flow (kgmole/h)

CH4 from absorber 37.78 3617.65 249.04
CO2 from regenerator 43.41 3534.92 173.77
Rich Amine mix 65.73 3548.70 2348.87
H2 from separator 66.21 482.63 1.16
H2S from regenerator 43.33 241.32 77.64
Regen Bottoms 131.56 262.00 2270.07
Inlet to Exchanger 66.21 482.63 2347.71
Inlet to Cooler HE-1 107.30 3617.68 2270.07
Lean Amine mix 43.33 3617.68 2273.60
Regen Feed 93.33 448.16 2347.71
Inlet to Separator 66.21 482.63 2348.87
Inlet to Pump 106.78 227.53 2270.07
Recycle 43.33 3617.68 2270.07

for a faster/stable convergence. Condenser type = total condenser. Y02 = Mole fractions of CO2 and H2S in the exit/treated gas respec-
The solver method = Modified HYSIM inside-out. Condenser tively, X1, X2 = Mole fractions of CO2 and H2S in the lean hybrid
pressure = 3617.68 kPa with DP of 15 kPa, reboiler pressure = 482. mix respectively, X01 , X02 = Mole fractions of CO2 and H2S in the rich
63 kPa, tray 1 temperature = 100 °C, reboiler temperature = 131.56 hybrid mix respectively. According to Seader and Henley [40],
°C. Reflux ratio = 1.5. Condenser exit temperature = 43.3 °C with heat across a stage/plate in the absorber (Fig. 3), the total material bal-
duty = 1.3  107 kJ/h. A mixer was used to combine the lean amine- ance is given by (1):
hybrid from the regenerator and make-up amine-hybrid mix. Mixer
G1 þ L1 ¼ G2 þ L2 ð1Þ
temperature and hybrid-mix mole number = 25 °C and 1 respec-
tively. Volume flow of cooled DEA-hybrid mix = 43 m3/h. DP in the
Or G1
G2 ¼ L2
L1 ð2Þ
condenser = 35 kPa. A pump was used to transport/recycle the
cooled DEA. To ensure steady state, the multiplier was set to 1 and where G1, G2 = Feed and treated gas flow rates, L1 and L2 = Lean and
the offset was set at
43.3 kPa. Number of parameters = ‘‘10” while Rich hybrid flow rates respectively. Then, for the contaminated gas
the transfer direction was chosen as ‘‘forwards” for all. The calcu- entering and leaving a stage,
lated and assumed stream values were compared within the speci- 0 0
G1 ðY 1 þ Y 2 Þ ¼ G2 ðY 1 þ Y 2 Þ ð3Þ
fied tolerance until they converged.
 0 0

G1 Y1 þ Y2
3. Theory: Material balance around the absorber and estimation ¼ ð4Þ
G2 ðY 1 þ Y 2 Þ
of absorption efficiency of the hybrid mix
For the components in the liquid entering and leaving a stage,
Considering the liquid–gas extraction system given in Fig. 3: Y1, 0 0
0 ¼ L2 X 1 þ L2 X 2 ð5Þ
Y2 = Mole fractions of CO2 and H2S in the feed gas respectively, Y10 ,

Please cite this article as: S. Eshorame Sanni, O. Agboola, O. Fagbiele et al., Optimization of natural gas treatment for the removal of CO2 and H2S in a novel
alkaline-DEA hybrid scrubber, Egyptian Journal of Petroleum, https://doi.org/10.1016/j.ejpe.2019.11.003
 S. Eshorame Sanni et al. / Egyptian Journal of Petroleum xxx (xxxx) xxx 5

Fig. 3. Typical Absorber operation with in and outgoing streams.

From (3) and (5), @U; @V and V are the change internal energy, change in volume
0 0 0 0 and volume of the gas respectively.
G1 Y 1 þ G1 Y 2
G2 Y 1
G2 Y 2 ¼ L2 X 1 þ L2 X 2 ð6Þ  
@H
@H ¼ C P @T þ @P ð18Þ
But from (2), @P T


G1 ¼ G2 þ L2
L1 ð7Þ @H; @T; @H
@P T;
@P; C p are changes in enthalpy, temperature,
Substituting (6) in (7) gives: enthalpy with pressure at constant temperature, and the specific
0 0
heat capacity at constant pressure. By interpolation, Eq. (18) was
G2 Y 1 þ L2 Y 1
L1 Y 1 þ G2 Y 2 þ L2 Y 2
L1 Y 2
G2 Y 1
G2 Y 2 used to estimate the heat duties in the pump, cooler and heat
0 0
¼ L2 X 1 þ L2 X 2 ð8Þ exchanger.

 
0 0 0 0 4. Results and discussion
G2 Y 1
Y 1 þ ðL2
L1 ÞY 1 þ G ðY 2
Y 2 Þ ¼ L2 ðX 1 þ X 2 Þ ð9Þ
2

 0
  0
 The results for the natural gas characterization are as given in
Y1
Y1 Y 1 ðL2
L1 Þ Y2
Y2
L2 Table 1a. The data in Table 1b is an excerpt of Table 1a that was
0 0 þ 0 0 þ 0 0 ¼ ð10Þ used in carrying out the simulation. Since the entire simulation
ðX 1 þ X 2 Þ G2 ðX 1 þ X 2 Þ ðX 1 þ X 2 Þ G 2
work was targeted at three major components, Table 1b was
From (2), G1
G2 ¼ L2
L1 , then obtained with its approximated data showing the actual concen-
 0
  0
 trations of methane, CO2 and H2S inputted to the Graphical User
Y1
Y1 Y 1 ðG1
G2 Þ Y2
Y2 L2 Interface/simulation environment of HYSYS during simulation
0 0 þ 0 0 þ 0 0 ¼ ð11Þ while other components were assumed to have zero concentra-
ðX 1 þ X 2 Þ G2 ðX 1 þ X 2 Þ ðX 1 þ X 2 Þ G2
tions owing to their insignificant concentrations.
Which gives: Tables 2a and b are plant data (Production and Exploration
      L Company, Nigeria) obtained from the gas treatment process for 4
1 0 G1 0 2
0 0 Y 1
Y 1 þ Y 1
1 þ Y 2
Y 2 ¼ ð12Þ samples. From Table 1, gas samples 1–4 are similar in components
ðX 1 þ X 2 Þ G2 G2
but differ slightly in composition hence, the samples can be seen as
     L same gas/one NG specie (NG with different compositions) since gas
0 G1 0 2 0 0
Y1 ¼ Y1 þ Y1
1 þ Y 2
Y 2
ðX 1 þ X 2 Þ ð13Þ samples from separate fields within a province have similar his-
G2 G2 tory; 10 kg each of the gas samples 1, 2, 3 and 4 were retagged
The absorption efficiency of the hybrid mix is then given by (14) A, B, C and D and treated with 10, 20, 30 and 35% DEA hybrid solu-
  tions respectively. The gases were delivered at different pressures
Initial amount of acid gas in feed
residual acid gas after absorption
as given in Table 2a. The molar flow rates of the absorbed acid
Initial amount of acid gas in feed
gases (i.e. CO2 and H2S) are not the same (Table 2b) due to their
 100% dissimilar molar masses.
ð14Þ For samples 1–4 (i.e. A-D), under the flow conditions estab-
Energy balance: lished in Table 2b (plant data), the amount of H2S in the feed gas
The energy of the system is estimated in terms of enthalpy was in the range of 52.8–58.8 ppm (Table 2a). After treatment with
changes which is a function of temperature and pressure of the DEA-hybrid mix, H2S concentration dropped to 1.08, 1.21, 1.09 and
absorber. 0.86 ppm in the treated samples respectively. For CO2, the initial
mole fraction in the feed gases A, B, C and D, were 0.18, 0.24,
H ¼ U þ PV ð15Þ 0.22 and 0.18 respectively, but after scrubbing, the CO2 mole frac-
tion reduced to 0.07, 0.11, 0.05 and 0.05, respectively. At increased
@H ¼ @U þ P@V þ V@P ð16Þ pressures, the removal of CO2 and H2S improved significantly.
H = Enthalpy, U = internal energy of the gas and PV = pressure * Based on the results in Table 2a, gas samples C and D gave low
volume which is a measure of the work done by the absorber and and medium pumping requirements of 830 and 1027.8 kscf/h
can be expressed as: respectively; the least residual CO2 concentration of 0.05 mol frac-
tion was obtained for gas samples (C and D) with corresponding
H ¼ f ðT; PÞ at constant v olume ð17Þ
H2S concentrations of 1.09 and 0.86 ppm respectively. The absor-

Please cite this article as: S. Eshorame Sanni, O. Agboola, O. Fagbiele et al., Optimization of natural gas treatment for the removal of CO2 and H2S in a novel
alkaline-DEA hybrid scrubber, Egyptian Journal of Petroleum, https://doi.org/10.1016/j.ejpe.2019.11.003
6 S. Eshorame Sanni et al. / Egyptian Journal of Petroleum xxx (xxxx) xxx

ber pressure was in the range of 3549–3535 kPa. The separator
temperature and pressure were estimated to be 66.2 °C and
482.6 kPa. The amine-hybrid regenerator temperature and pres-
sure were in the range of 93.3 – 131.6 °C and 262–448.16 kPa,
respectively. H2S and CO2 have dissimilar molar masses of 34
and 44 kgmole-1 respectively, hence their gas velocities are not
the same. The gas flow rate for NG was in the range of 830 –
2203.1 kscf/h and the DEA-hybrid molar flow rate and temperature
were 251.8 kgmole/h and 52.8 °C respectively. Considering Table 3,
the results from pre-gas-treatment and post-gas-treatment opera-
tions for the 4 gas samples (Table 2b) are given for 10–35% DEA-
alkaline mix and the residual CO2 and H2S in the treated samples
1–4 are in the range of 0.05–0.11 and 0.86–1.21 ppm after pre-
treatment based on plant data which give accuracies of 61, 54,
77, 72% for CO2 removal and 98, 98, 98, 99% for H2S removal,
respectively. For higher %DEA in the hybrid mix, absorption of
CO2 and H2S increased owing to the additional properties infused
in the hybrid DEA mixtures for CO2 and H2S dissolution. Comparing
plant with simulation results for sample 4 (i.e. pre- and post-
treatment plant data in Table 2b with simulation data in
Table 4a), the results from plant operations and simulation are in
close agreement but, the simulation results for CO2 gave higher
accuracies/removal rates which may have resulted from changes
in pressure, flow rate, temperature and concentration of absorbent
[39]; gas sample 4 was chosen because, the best acid gas removal
was obtained for the sample. Table 3 gives the pressure loss of the
incoming stream at the pump inlet/suction and exchanger as
3390 kPa and 34.47 kPa respectively. The stream temperature
was cooled to
63.96 °C in the cooler after going through the
exchanger. The power consumptions in the three equipment imply
that the cooler performed more work than the heat exchanger and
pump which further implies that a substantial amount of heat was
dissipated in the cooler. The adiabatic efficiency of the pump is 75%
which may have resulted from pressure or temperature drop at
pump casing thus reducing the flow rate of the pumped liquid
[39]. The molar flow rates of the stripped CO2 and H2S are given
in Table 3. Furthermore, only 6% of the reboiler heat was lost in
the absorber which makes energy conservation in this unit about
94% efficient.
Concentrations of acid gas components in the Ca(OH)2-amine
stream were expressed in terms of loadings i.e. ratio of moles of
acid gas to moles of the Ca(OH)2-amine mixture. The spreadsheet
in Aspen HYSYS was used to estimate the loadings directly which
were then incorporated in the simulation to serve as control for
optimizing the gas treatment process and the results are as given
in Fig. 4. Conditions in the absorber, separator and regenerator
are as given in Table 3. In order to pass a good judgement for the
treated gas samples, gas sample 4 which contained the highest
amount of CO2 and H2S, was used to determine the best treatment
conditions for the gas samples (see Tables 4 a and b).
The results obtained from the control experiments discussed in
section 2.1 are contained in Table 4b; efforts were made to main-

Table 4a
Summary of Simulation Results for Treated Gas Sample 4 (Sample D).

Parameter Simulation Data for Treatment of Sample 4 Data for Repeated Simulation for gas sample 4
Case1 Case2 Case3 Case4 Case5 Case6 Case30 Case40 Case 7 Case 8
(10% DEAhyb) (20% DEAhyb) (30% DEAhyb) (35% DEAhyb) (25.3% DEAhyb) (27.4% DEAhyb) (30% DEAhyb) (35% DEAhyb) (40% DEAhyb) (50% DEAhyb)
Gas Flow (kscf/h) 1024.58 2203.1 1024.58 1027.8 2203.1 830 1024.58 1027.8 2203.1 2203.1
Ca(OH)2-DEA Flow 251.8 251.8 251.8 251.8 251.8 251.8 251.8 251.8 251.8 251.8
(kgmole/h)
T (⁰C) 52.8 52.8 52.8 52.8 52.8 52.8 52.8 52.8 52.8 52.8
P (kPa) 5000 5000 5310 6040 5270 5250 5310 6040 5000 5000
Valve Opening – – – – – – – – – –
Mole frac CO₂ (-) 0.16 0.024 0.00 0.001 0.004 0.00 0.00 0.00072 0.00 0.00
ppm H₂S (ppm) 12.19 6.28 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Mole frac NOx (-) – – – – – – – – – –
Mole frac CN (-) – – – – – – – – – –

Fig. 4. Simulation results of CO2 and H2S absorption from NG using 0.1 M, 5 w/w% Ca(OH)2 and 10–50% DEA.

Please cite this article as: S. Eshorame Sanni, O. Agboola, O. Fagbiele et al., Optimization of natural gas treatment for the removal of CO2 and H2S in a novel
alkaline-DEA hybrid scrubber, Egyptian Journal of Petroleum, https://doi.org/10.1016/j.ejpe.2019.11.003
 S. Eshorame Sanni et al. / Egyptian Journal of Petroleum xxx (xxxx) xxx 7

tain the same process conditions for each solvent (i.e. DEA only or
Ca(OH2) only) used in the gas cleaning process as contained in
Table 4a for 10, 20, 30 and 35% concentrations of the hybrid mix.
Here, the DEA solution was able to trap CO2 and reduce the CO2
gas concentration to 0.03 at increased concentrations of DEA from
10 to 30%, however, at 35% DEA concentration, the amount of
trapped CO2 increased for 10–30% DEA but increased slightly by
about 27% for 35% DEA over the value obtained for 30% DEA solu-
tion; this may have been caused by the resistance posed by the
presence of some H2S in the DEA after some time. This can be jus-
tified by the results in Table 4b where more H2S was absorbed by
the DEA at increased concentration, thus, this also suggests an
optimum around DEA concentration of 30%. In addition, the Ca
(OH)2 solution reduced the H2S and CO2 concentrations of the
gas significantly (i.e. from 58.8 to 0.003 for H2S and from 0.18 to
0.042 mol fraction for CO2) however, in lieu of the DEA solution
being able to reduce the CO2 and H2S significantly, there was con-
tinuous reduction in CO2 concentration in the gas whereas, for H2S,
the best results were obtained with 10% Ca(OH)2 which also
defines an optimum concentration around the 10% for Ca(OH)2.
This may have been caused by steric effect of the overlapping ions
from the solution which form a repulsive cloud that tends to pre-
vent further absorption or interaction of electron species between
the H2S gas and the Ca(OH)2 solution at 20–30% concentration (i.e.
see Table 4b; 11.6 ppm is the concentration of H2S in the gas after
treating with 20 and 30% Ca(OH)2 solution). However, increasing
the solution concentration of Ca(OH)2 to 35% breaks this electron
cloud and thus allows for further transfer/dissolution of the H2S
molecules into the Ca(OH)2, thus giving H2S residual gas concen-
tration of 8.4 ppm. Overall, the CO2 concentration in the gas
decreased for the 10–30% DEA wt/wt solution but increased for
the 35% DEA solution; this then suggests that, the 35% DEA solution
must have been too saturated with water which mixes with the
DEA and would not allow further comingling of the DEA and water
molecules. Beyond 20% DEA conc i.e. at 30% DEA wt/wt%, the conc
of CO2 in the gas increased to 0.06 from 0.033. Also, for the DEA
solution, H2S concentration only dropped to zero for the 30–35%
DEA solution which suggests that the optimum DEA concentration
for H2S removal can be as high as 35% or more when used alone for
natural gas treatment, hence, the need to consider optimizing the
gas treatment process in order to obtain better results. Comparing
the results obtained in Table 4a with that of Table 4b, it is clear that
better removal rates were obtained.
The results for the quantities of CO2 and H2S removed from
sample 4 (sample D) as well as the estimated efficiencies have been
summarized in Tables 4a and b, and 5 respectively. Based on the
results, if scrubbing efficiencies >90% are desired, DEA-hybrid solu-
tions of 10–20% concentrations should not be used for scrubbing
NG contaminated with similar compositions of H2S and CO2
(Table 1b) since both solutions gave the lowest scrubbing efficien-
cies from experiment and simulation studies (Table 5).
Considering the highest and minimum pumping rates/gas flow
rates of the gas, the flow rates of the simulated gas treatment oper-
ation was adjusted (Table 4a) to fall within the flow rates adopted
for 10–35% DEA-hybrid solution as given in Table 2a. In Table 4, the
results for CO2 and H2S removal imply that, gas treatment with
concentrations in the range of 27.4–30% DEA hybrid solution gave
the best result. Also, the results suggest that treating the gas with
10% DEA hybrid mix is not favourable considering the estimated
percent efficiencies i.e. < 40%. Maximum amine-hybrid loading
for removal of CO2 and H2S as specified by Peng and Robinson
are 0.45 and 0.35 respectively [41], whereas, this work gives better
Ca(OH)2-DEA and lower mixture concentrations of 0.27–0.3 for the
removal of both gases from natural gas (Table 4a, Fig. 4).
For gas sample 4, between 2 and 2.7 bar, the CO2 and H2S gas
concentrations in the treated gas were negligible. Gas treatment
at such pressures leaves the stream almost entirely free of acid
gases whereas, for pressures within 3–4 bar, CO2 removal seems

Table 4b
Summary of Simulation Results for Treated Gas Sample 4 (Sample D).

Parameter Experimental Data for Sample 4-DEA only Experimental Data for Sample 4-Ca(OH)2 only
Case1 10% Case2 20% Case3 30% Case4 35% Case1 10% Case2 20% Case3 30% Case4 35%
Gas Flow (kscf/h) 2203.1 2203.1 1024.58 1027.8 2203.1 2203.1 1024.58 1027.8
LiqFlow (kgmole/h) 251.8 251.8 251.8 251.8 251.8 251.8 251.8 251.8
T (⁰C) 52.8 52.8 52.8 52.8 52.8 52.8 52.8 52.8
P (kPa) 5000 5000 5310 6040 5000 5000 5310 6040
Valve Opening 4/5 4/5 3/5 3/5 4/5 4/5 3/5 3/5
Mole frac CO₂ (-) 0.17 0.06 0.033 0.042 0.087 0.064 0.048 0.03
ppm H₂S (ppm) 20.2 15.5 5.88 0.003 7.6 11.6 11.6 8.4
Mole frac NOx (-) – – – – – – – –
Mole frac CN (-) – – – – – – – –

Table 5
Scrubbing Efficiencies for the Alkaline-DEA Hybrid Solutions.

Samples 1–4 Sample 4

H2S Cnc CO2 Cnc CO2 removal eff H2S removal eff H2S Cnc CO2 Cnc %DEA HYBRID CO2 eff H2S eff
S1 58.2 0.18 61 98 58.8 0.24 10 33 79
S2 58.8 0.24 54 98 20 90 89
– – 25.3 100 100
– – 27.4 99.6 100
S3 58.5 0.22 77 98 30 98.3 100
S4 58.8 0.18 72 99 35 100 100
40 100 100
50 100 100

Hint:
Cnc = Concentration, eff = efficiency, S1-4 = Samples 1–4.

Please cite this article as: S. Eshorame Sanni, O. Agboola, O. Fagbiele et al., Optimization of natural gas treatment for the removal of CO2 and H2S in a novel
alkaline-DEA hybrid scrubber, Egyptian Journal of Petroleum, https://doi.org/10.1016/j.ejpe.2019.11.003
8 S. Eshorame Sanni et al. / Egyptian Journal of Petroleum xxx (xxxx) xxx

10-35% DEA mix HYSYS H2S 10-35% DEA hybrid mix (exp.) H2S
10-35% DEA mix HYSYS CO2 10-35% DEA mix exp. CO2

10

8
Gas Composition (ppm)

0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5

-2

Operating Pressure (bar)

Fig. 5. Mole fraction of CO2 and H2S in the treated gas sample 4 at different pressures.

to be higher compared to H2S (Fig. 5) hence, there is a close agree-
ment between plant data and simulation results.
Another reason for the higher accuracy of CO2 removal relative
to H2S is the increase in pressure [42] above the critical value, say
in the region of 5310 kPa with corresponding flow rate of 1024.58
kscf/h where acid gas removal is highest (see left side of
Table 4a); the reverse is the case at higher or lower pressures
and concentrations, which then suggests the need to establish
an optimum point as given on the right side of Table 4a. Consid-
ering [39], for non-hybrid DEA, the required concentration of
amine/absorbent lies between 0.35 and 0.65 mol amine per mol
acid gas absorbed. However, results from simulation were best
for DEA-hybrid mix with concentration range of 27.4–30% mol
amine/%mol gas which reduces the cost of the amine used; this
also validates the findings of Erfani et al. [6] and Van-Daal and
Bouallou [43].
Furthermore, the 27.3% DEA-hybrid mix gave the best results
due to changes in pressure, flow rate [42,44] and the concentration
of the absorbent used [39].
Considering the results of other modelling and simulation
works [28–31,35], where scrubbing efficiencies <98% were
obtained, and the findings of this study, 100% acid-gas removal
efficiency can be guaranteed (Table 5).
4.1. Effect of pressure and temperature on the absorption/process
economics
While should be noted that feed conditions remain unchanged
in an absorption process, interstage variation in process conditions,
such as flow rate, pressure and temperature occur because of the
counter-current flow of the feed gas (natural gas) and the hybrid
absorbent. Hence, it becomes necessary to account for these

Please cite this article as: S. Eshorame Sanni, O. Agboola, O. Fagbiele et al., Optimization of natural gas treatment for the removal of CO2 and H2S in a novel
alkaline-DEA hybrid scrubber, Egyptian Journal of Petroleum, https://doi.org/10.1016/j.ejpe.2019.11.003
 S. Eshorame Sanni et al. / Egyptian Journal of Petroleum xxx (xxxx) xxx
changes. Across the absorber, inter stage variation in pressure,
temperature and flow rate are caused by the interactions between
absorbent and the absorbate which bring about pressure drop in
the gas stream caused by heat exchange/mass transfer of mole-
cules between the absorbent and the natural gas. In the acid gas
removal process described by Al-Marzouqi et al. [45], for different
absorbent-concentrations at increased pressures, the solubility of
the acid gases (H2S and CO2) increased when the %H2S removed
increased with pressure for 1 wt% DEA+(10–30)wt% K2CO3 solu-
tions. However, at 50 bar, the highest %H2S removed was 99% for
the DEA + 20–30 wt% K2CO3 solution, whereas, it was 99.3% for
the 10 wt% K2CO3 + 1 wt% DEA, hence, the solution with 10 wt%
base gave the best H2S removal with a slight difference of 0.3%.
However, the CO2 removal rates were 97, 85.7 and 73.8% for the
1 wt% DEA + 30, 20 and 10 wt% K2CO3 solutions respectively. This
confirms the trend obtained in the current study despite the differ-
ence in types of solvent mixed with DEA (i.e. K2CO3 – a base and Ca
(OH)2 – an alkali) in their work relative to the current study. Based
on the trend observed, they suggested the need to determine the
optimum pressure for highest/maximum gas removal so as to limit
or reduce pumping costs which will save some amount of energy,
as well as the amount of time spent on recycling the treated gas
when it falls below the required specification at conditions below
the optimum pressure. Also, the integrity of the equipment is pre-
served since equipment hammering/bombardment could ensue if
the optimum pressure is exceeded; excessive pressures can result
in hazards or explosions, loss of absorbent and low CO2/H2S
removal efficiencies, which may be quite expensive to manage
considering the cost that may accrue from repairs, replacement,
lost production/processing time and recycling to meet specifica-
tion. They also recounted that, the effect of altering the feed gas
pressure is not pronounced hence, regulating the feed gas flow
rate/pressure may not be necessary in an absorption process. In
lieu of this, regulation of the process or inter-stage pressures dur-
ing absorption is very important. Also, in their work, even when
the flow rate of absorbent was constant, the effect of temperature
was seen for water absorption of H2S and CO2 at 65 and 100 °C
with the gas temperature fixed at 50 °C. It should be noted here
that the gas temperature is fixed but the liquid/water temperature
was varied. For both acid gases i.e. H2S and CO2, absorption in
water was lower at higher temperatures and vice-versa. This is
because, at 100 °C which is the boiling point of water, the gas
molecules are activated and their average kinetic energy is
increased which makes the absorbate-molecules travel faster
through the absorbent thus reducing the contact time. The longer
the contact time between the absorbate and absorbent, the better
the absorption and vice-versa for both physical absorption and
chemical absorption. Again, at an optimum pressure of 50 bar,
the best acid gas removal was achieved. Based on their results,
inter-stage temperature effect is significant when the absorbent
is either water or base i.e. K2CO3/alkali. However, if DEA solution
is used alone, the effect of temperature on the absorption process
is less significant and this may be due to the slow response of
the viscosity of DEA to temperature changes, hence, it is advisable
to use DEA of low/measurable concentrations alongside other sol-
vents for effective natural gas purification. Other drawbacks on the
use of DEA solution as the only absorbent for acid gases include
equipment corrosion, solvent degradation, foam formation and
high energy requirements (due to its viscosity) hence, the need
to consider combining DEA with other suitable solvents such that
both adsorbents offer their individual characteristics/synergistic
effects in the combined adsorbent. This also would help reduce
the cost of DEA solution needed for use with cheaper solvents that
have good absorption potentials for acid gases. In addition, in the
same study, the absorption rate increased with increase in temper-
ature for the absorbent (DEA + K2CO3). This is because, K2CO3 is
Please cite this article as: S. Eshorame Sanni, O. Agboola, O. Fagbiele et al., Optimization of natural gas treatment for the removal of CO2 and H2S in a novel
alkaline-DEA hybrid scrubber, Egyptian Journal of Petroleum, https://doi.org/10.1016/j.ejpe.2019.11.003
9
lighter than DEA and can easily get heated and transported with
the gas at higher temperatures.
The work of Pouladi et al. [46] also considered the effect of tem-
perature and pressure on the absorption of CO2 from ethane gas
using DEA only, or pure DEA as absorbent. The design efficiency
of the plant was estimated to be a CO2 capture of 85% which is
equivalent to 19 ppm in the absorber outlet. The data generated
was simulated using ASPEN HYSYS. They employed a static
approach in determining the optimum conditions and found that
pressure variation in the system did not have any significant effect
on the absorption process. This may have been caused by the use of
DEA only, its viscosity as well as the low pressure variation/differ-
ence i.e. 24–26 bar adopted in their study. However, it was
observed that, the absorption efficiency of the amine improved
with increase in amine-volumetric flow rate and the absorption
tower inlet temperature. The effect of temperature was highly sig-
nificant as obtained from simulation studies. The optimum condi-
tions for the CO2 absorption from ethane as feed gas were found to
be 2118 kg mol h
1, 75 m3 h
1, 53 °C and 24 bar, which are indica-
tive of the molar flow rate of amine, the volumetric flow rate of the
amine as well as the system’s temperature and pressure, respec-
tively. Again, because DEA is highly viscous, the pressure required
when used alone is usually high if high acid gas removal efficien-
cies are desired. They obtained an optimum pressure of 50 bar
which is way higher than the pressure obtained in the current
study. The increase in absorption efficiency was observed from
30 to 53 °C but beyond and below 53–70 °C, CO2 absorption
declined. From all observations, these then suggests that, the beha-
viour exhibited by each system is a function of the absorbent, the
absorbate/target component and the gas to be treated. It was also
observed that the amine flow rate dropped from 120 m3 h
1 to
75 m3 h
1 at optimum conditions which then informs a low circu-
lation/pumping cost of the amine; in essence, the results corrobo-
rate the findings in the current study. Although, pressure variation
may not be somewhat significant for lone amine (i.e. when only
amine is used as absorbent), however, changes may then set in if
the system is hybridized (i.e. one involving combined absorbents)
such that, the drawbacks of the amine (i.e. high viscosity, degrada-
tion, corrosion etc.) are complemented by the additional solvent.
Furthermore, there exists a mutual relationship between the pres-
sure and temperature exerted by a gas having a close behaviour to
that of an ideal gas i.e. its pressure decreases with temperature
when the gas’ behaviour is near/approximately that of an ideal
gas. This implies that, DEA absorbent, when used alone, renders
the ideal-effect of such gases negligible.
4.2. Effect of temperature on the raw gas and the lean Amine-Hybrid
Mix: The rule of thumb
Several rules are employed in the design and operation of gas
sweetening units. These rules were enacted several years back.
The question often arises as to how absolute or flexible these rules
are. Several of these rules have been suggested for absorber condi-
tions. The rules specified include a minimum absorber temperature
of 5 °C (i.e. the temperature differential between the incoming acid
gas and the lean amine-hybrid feed) in the absorber, 0.12 kg/L
specification for reboiler steam, 99 °C lean/rich exchanger outlet
temperature, as well as the regenerator pressure/reboiler temper-
ature. Although these conditions or ‘‘rules of thumb” are excellent
starting points, however, none of them actually depicts the opti-
mum conditions for all cases and, considering the peculiarity of
the situation, violating these rules could in turn offer considerable
advantages/disadvantages to process efficiency [47], especially
when the circulating fluid/coolant is a refrigerant. Gas composi-
tions differ, and several situations may require a thorough investi-
gation as to the validity of these set points and whether they are
10 S. Eshorame Sanni et al. / Egyptian Journal of Petroleum xxx (xxxx) xxx
beneficial to the entire process as well as offset any additional
risks. Maintaining the minimum temperature of 5 °C in the absor-
ber helps to prevent condensation of any natural gas liquid in the
hydrocarbon, which will in turn lower the performance of the
absorber due to foaming, loss of product and blocking of carbon fil-
ters or pipelines [25,48,49]. However, cases of natural gas liquid
(i.e. components  C2) condensation have been recorded at tem-
peratures <28 °C for the Khazak type of natural gas [47]. Also, it
has been established that, decreasing the temperature of the absor-
ber improves its performance, which in turn enhances the absorp-
tion of H2S while CO2 absorption is more favoured at higher
temperatures (i.e. 30–50 °C) and vice-versa [47], thus, in the cur-
rent study, the amine-hybrid alkaline mix helps to alleviate this
problem as the Ca(OH)2 helps to complement the mix by modify-
ing the amine component of the hybrid absorbent such that, its
affinity for CO2 is enhanced despite the temperature reduction
effect which may seem to favour more H2S absorption relative to
CO2. Furthermore, studies have shown that, at temperatures in
the range of 5–28 °C, CO2 absorption is low while that of H2S is
higher and vice versa. In addition, it was observed that high tray
temperatures at the bottom of the absorber minimized any reduc-
tion in the kinetic rate of CO2 absorption. According to them, it is
advisable to keep the bottom tray temperature of the absorber at
30–40 °C above the feed gas temperature, as this helps to create
near equilibrium conditions. This further suggests that, practical
limitations of the scrubber/absorber should not be ignored. For
example, temperatures <25 °C have been reported to cause a dras-
tic increase in DEA viscosity thus making the operation of the con-
tactor/absorber/scrubber exceedingly difficult. Also, high
temperatures can also speed up amine degradation. In this study,
the inlet stream to the exchanger is at 66.21 °C and reaches the
pump at 106.78 °C before entering the cooler at 107.30 °C (see
Table 3); the inlet stream temperature to the cooler is quite exces-
sive and will therefore lead to undesirable consequences in the
absorber [47] hence, the need to cool the stream before it enters
the absorber. In the current study, operating at an absorber tem-
perature of 37.8 (methane exit temperature at the top of the absor-
ber) to 65.73 °C (rich amine-hybrid mix temperature at the
absorber bottom) also helps limit the tendencies for the recovery
of inherent natural gas liquids (NGLs or condensates) [47–49].
Again, looking at Table 3 of the current study, the lean amine
mix to the absorber enters at 43.3 °C which is within the recom-
mended temperature spec of 45 °C [47].
For clarification, a closer look at Fig. 2 shows the heated lean
amine-hybrid mix which leaves the regenerator and transits via a
pump through HE-1. At the start of the operation, fresh amine-
hybrid mix is taken at the suction of the pump linked to line Q3.
At this point in time, before entering the absorber, the lean
DEA + Ca(OH)2 stream then goes to the cooler (second heat exchan-
ger) which uses cooling water to help moderate the temperature of
the regenerated amine-hybrid mix before it is sent to the absorber
or storage unit after the entire process. In order to avoid thermal
shock, once the incoming lean amine-hybrid mix is certified free
of acid gases at the sample point on the line, its temperature is
moderated before it enters the absorber, but if otherwise, the lean
amine-hybrid mix is sent back/recycled to the regenerator for
regeneration/stripping of the acid gases, and then the cycle goes
on. At this time if the lean amine mix is certified okay, it is then
allowed to flow to the absorber through the heat exchanger which
serves to cool the lean amine entering the absorber; the process is
largely a physical process and, heat sensitive, especially at the
amine-hybrid mix regeneration stage (desorption stage). The rea-
son for cooling this lean amine stream is to help regulate the tem-
perature of the absorbent-mix prior its entry to the absorber so as
to avoid the tendency for thermal shock i.e. if this stream is not
cooled, there may be a sharp or sudden rise in temperature of
Please cite this article as: S. Eshorame Sanni, O. Agboola, O. Fagbiele et al., Optimization of natural gas treatment for the removal of CO2 and H2S in a novel
alkaline-DEA hybrid scrubber, Egyptian Journal of Petroleum, https://doi.org/10.1016/j.ejpe.2019.11.003
the NG or sudden drop in the amine-hybrid mix temperature
which may cause the escape of some of the acid gases or loss of sol-
vent due to the less contact time available for the lean amine-
hybrid mix and the NG as they flow counter-currently in the absor-
ber; also, there may be some evidential equipment hammering.
Thus, moderating the temperature of the amine-hybrid mix via
cooling is necessary in order to maximize or attain high efficiency
in the absorption process which in turn affects the gas-amine
hybrid interaction during acid gas stripping from the NG. Another
function of the cooler is that it would help cool the regenerated
DEA-alkaline hybrid mix upon complete removal of the acid gases
before the mix is sent to the storage unit for future use. Also, since
the amine-hybrid mix is less viscous than the ordinary amine, it is
necessary to transport the mix at moderate temperatures in order
to abate the tendencies for thermal degradation of the DEA-hybrid
mix. For practical purposes, according to Addington and Ness [50],
it is somewhat easy to control the absorber temperature while
bearing in mind the feed gas temperature because, there are evi-
dences of hydrocarbon dew point variations with respect to natural
gas compositions. However, a blanket rule that bothers on the feed
temperature set point may cause the operator to lose out on some
benefits that may accrue from lowering the cooler temperature.
Furthermore, despite the potential for saving operating costs, a
decrease/increase in absorber performance may result in bene-
fits/loses when the feed gas temperature decreases/increases due
to capacity limitations imposed by operational constraints on the
absorber. However, since absorptions of this type are quite pecu-
liar, an investigation into optimizing the system’s performance is
necessary in order to maximize the benefits and risks of varying/-
controlling the lean amine temperature [50].
4.3. Economic effect of optimizing pressure and temperature in a gas
cleaning process
To process larger quantities of gas, more stringent specifications
ought to be achieved by operating profitably at optimal conditions.
This then gave rise to the need for searching other viable alterna-
tives. From the study of Lunsford and Mcintyre [51], for chemical
systems involving ethanol amines, potassium carbonate or an
alkali, an increase in H2S absorption or decrease in CO2 absorption
may ensue based on the ‘sorption kinetics. This may also slightly
enhance the absorption of alkanes greater than C1 such as C2-C7
while ensuring efficient dehydration of the treated gas. According
to them, for physical and chemical solvents that may approach
their equilibrium concentrations, operating the absorber at lower
temperatures may promote the uptake of light contaminants.
However, considering the case of chemical solvents, whose reac-
tions may have some rate/kinetically limiting steps, colder/lower
temperatures may subsequently result in lowering the absorbed
quantity of a particular contaminant. They also asserted that, an
attempt to efficiently and economically eliminate H2O, H2S and
CO2 helps to reduce the tendencies for hydrate formation, environ-
mental pollution etc. They added that, solvent type, concentration,
pressure, and solvent circulation can be manipulated to produce
high quality hydrocarbon product specs. Generally, based on the
vapour-liquid equilibrium of absorption systems, lower tempera-
tures tend to improve absorption of light molecular weight compo-
nents. Also, it has been stated that physical solvents obey this
principle, whereas, for an absorption process that involves some
measure of reaction/reactive interactions, this may not be true
for systems that may approach equilibrium, hence, lower temper-
atures may not favour the absorption of low molecular weight
components. But, in the event that the absorption is kinetically
limited i.e. the case with CO2 in contact with certain amines, tem-
perature effect on the absorption process will require certain addi-
tional information for the effect to be explicit. This anomaly/
 S. Eshorame Sanni et al. / Egyptian Journal of Petroleum xxx (xxxx) xxx
ambiguity comes from the permissible competing phenomena/
opposite effect temperature has on reaction rates and solubility.
Evidence also has it that, for H2S and CO2 absorption in alkanolami-
nes or mixtures of amines and other solvents, H2S absorption
reaches equilibrium while CO2 absorption is kinetically limited in
some instances. However, they concluded that, the performance
of various amines and physical solvents may be compared based
on criteria such as solvent and feed gas temperatures. Furthermore,
a clear insight on the conflicting phenomena of absorption-
equilibrium and kinetics may breed situations where optimum
gas cleaning conditions may be exploited for profitable operations
[51].
Basiri [52] stated that chemical absorption processes are gener-
ally categorized as systems highly dependent on heat of absorption
which also require a high heat of regeneration of the spent absor-
bents. Consequently, the quantity of steam supplied to a reboiler
increases as the heat of regeneration increases which also imposes
a resultant increase in production cost. The operation of a stripper
column in series with an absorber in an absorption process as done
in the current study, helps to reduce the heat spent in regenerating
the spent solvent which subsequently lowers the heat requirement
of the process thus enhancing the continuous removal of CO2 from
natural gas. Process parameters including concentration, pressure
and temperature were altered in order to determine the optimum
process conditions required for minimum heat supply in the regen-
erator. Based on the results obtained in [52], the optimum condi-
tions for optimal CO2 removal from natural gas are DEA
concentration, reboiler temperature and stripper pressure of
25–35 wt%, 118–121 °C, and 1.5–1.75 bar respectively. Despite
showing a variation in pressure, the pressure values obtained in
the study are not too far from the lower limit pressure obtained
in this study i.e. 2–2.7 bar; this may have been caused by variation
in other process parameters such as flow rate, type of absorbent,
operating temperature and regenerator conditions.
The study carried out by Aliabad et al. [53], for CO2 and H2S
removal using MDEA and DEA in an absorber with 20 stages, gave
amine outlet mass flow rates of 12938.94 and 19276.6 kg mol/h
which are both 51.4 and 76.55 times the mass flow rate (i.e.
251.8 kgmole/h; see Table 4) obtained in this study. Considering
the difference in the estimated flow rates for both studies, it then
suggests that the current study lowers the pumping cost signifi-
cantly, with high levels of accuracy in the range of 99–100% for
H2S and CO2 removal from natural gas.
4.4. Reasons for avoiding the use of Ca(OH)2 or DEA only for the
absorption of CO2 and H2S and its effect on the process economics
For ethanol amine, literature has it that, the heat of regenera-
tion is 20 Giga Joule per tonne-carbon (GJ/t-C) while for an alkali
such as NaOH/(CaOH)2 it is about 15 GJ/t-C. It is no arguable fact
that the recovery of a fluid in an absorption process is usually at
higher temperatures (high temperature swings)/lower pressures
(low pressure swings) than those required during absorption.
However, unlike NaOH, the recovery of Ca(OH)2 can be achieved
without the need for calcination at about 900 °C [54]. Although
the use of DEA for gas absorption is a well proven technology with
extensive commercial application as well as capital costs, however,
the solvent has low tolerance for SO2, NOx and O2 which are often
present in natural gas. Also, since DEA is usually administered in
low concentrations in order to abate issues such as equipment cor-
rosion, this also limits its loading capacity for CO2. In addition, DEA
can undergo degradation when contaminated. Compared to Ca
(OH)2, DEA attracts higher energy requirements during regenera-
tion of the spent solvent in an absorption process [54]. There is also
the issue of solvent loss due to mist-formation for absorption sys-
tems using DEA only. An evaluation of new amine-based solvents
Please cite this article as: S. Eshorame Sanni, O. Agboola, O. Fagbiele et al., Optimization of natural gas treatment for the removal of CO2 and H2S in a novel
alkaline-DEA hybrid scrubber, Egyptian Journal of Petroleum, https://doi.org/10.1016/j.ejpe.2019.11.003
11
such as those with: low heats of absorption, high CO2 loading
(i.e. strong bicarbonate formers such as calcium hydroxide),
reduced degradation/desorption and addition of additives/Ca
(OH)2 to improve absorptive performance bring about an increase
in reaction rate, CO2 loading and process improvements as well as
to lower energy requirements, improve heat integration, increase
solvent/DEA absorptive capacity, reduce in solvent loss (which
improves mass transfer for contact between the absorbate and
absorbent) as well as the ability to inhibit solvent/DEA degradation
which is one of the useful properties infused by Ca(OH)2 in the
hybrid amine mix.
Based on ref. [54], the following mechanisms were proposed for
the reactions between the acid gases and the DEA-Ca(OH)2 mix for
the current study:
R1 R2 -NH + CO2 ! R1 R2 -NH2 + R1 R2 -NCOO-
ð19Þ
þHeatðreactionequationforDEAonlyÞ
R1 R2 -NH + H2 S ! R1 R2 -NH2 + R1 R2 -NH2 -
ð20Þ
þS þ HeatðreactionequationforDEAonlyÞ
Ca(OH)2 + CO2 ! CaCO3 + H2 O (reaction equation for Ca(OH)2 only)
ð21Þ
Ca(OH)2 + H2 S ! Ca(HS)2 + 2H2 O (reaction equation for Ca(OH)2 only)
ð22Þ
Ca(OH)2 + R1 R2 -NH + CO2 + H2 S ! R1 R2 -NH2 + R1 R2 -NCOO-
þHeat þ SðreactionequationforDEAandðCaOHÞ2 combinedasabsorbentÞ
ð23Þ
In equations 1a and b and, 2a and b, although each solvent (i.e.
DEA or Ca(OH)2) has high affinity for H2S and CO2, however, the
CO2 and H2S in the NG will compete for the DEA or Ca(OH)2 solu-
tion thus making each solution give more preference to one type of
acid gas relative to the other when used separately; the absorption
preference of the acid gases with respect to DEA is more to H2S
than CO2 while, for the Ca(OH)2, the preference is more to CO2 than
H2S because, CO2 is polar with respect to the aqueous/water-
containing Ca(OH)2. The beauty of the combined mixture of DEA-
Ca(OH)2 system is that, unlike other systems where pH regulation
can only be done by the addition of acid such as HCl, the hybrid-
system described is self-regulating in terms of pH (i.e. from about
10.41) which favours the formation of CaCO3 to a pH of 8–10 (i.e.
8 < pH < 10) which favours the formation of Ca(HCO3)2 [55]; this
pH regulation helps to convert the insoluble CaCO3 formed to
CaHCO3 by an extra step that is brought about by the continuous
absorption of CO2 by the caustic alkali. Borrowing a leaf from the
explanation offered by Rackley [54] on the use of NaOH as absor-
bent for CO2 capture, the use of Ca(OH)2 as absorbent in the current
study is of great advantage because, for the case of NaOH, which is
a group 1 alkali, the NaOH regeneration is quite expensive because,
it cannot be achieved under mild conditions such as low tempera-
ture and pressure swings, hence it will require an extra chemical
reaction step such as the addition of Ca(OH)2 to the carbonate
(Na2CO3) formed from the trapped CO2 + NaOH reaction. The addi-
tion of CaCO3 to Na2CO3 then forms CaCO3 and NaOH. The CaCO3 in
this case is not in continuous contact with CO2 and hence the slurry
would have attained equilibrium and would need to be heated at
about 900 °C to drive off water and calcine the CaCO3 for CO2
regeneration; this actually makes the process more expensive
and a bit more complicated. Thus, considering the last statement,
for the case of the hybrid solvent comprising of Ca(OH)2 and
DEA, the produced water combines with some of the available
CO2 to form H2CO3 which helps to solubilize/dissolve the cloudy
12 S. Eshorame Sanni et al. / Egyptian Journal of Petroleum xxx (xxxx) xxx
CaCO3 obtained from the reaction between the Ca(OH)2 and CO2
leading to the formation of soluble calcium hydrogen carbonate
CaH(CO3)2 which is a clear solution. This continues up to/through
the quasi-equilibrium state and finally ceases when equilibrium
absorption of CO2 is attained. The formation of soluble CaH(CO3)2
makes the regeneration of the CO2 easy without any need for it
to be calcined at an extreme temperature of say 900 °C as high-
lighted for NaOH + DEA absorbent and, according to Chandel and
Williams [56], high cost involvement of CO2 capture during natural
gas synthetis corresponds to increased cost of natural gas obtained
from the process.
5. Conclusion
NG purification was achieved using DEA solutions of 27.4–30%
v/v with 0.1 M Ca(OH)2 flowing at 258.1 kg-mole/h and at
52.8 °C. The results imply that, the quantity of acid gas removed
is a measure of the quantity of Ca(OH)2-DEA mixture used in treat-
ing the NG. The simulation results were seen to conform with stan-
dard pipeline specifications of 2.0 vol% CO2 and 4 ppm (volume
H2S) which are indicative of zero tolerance for waste and environ-
mental pollution. From the results, the optimum pressure for treat-
ing the natural gas lies within from 2 to 2.7 bar. Therefore, an
optimization of the process parameters and absorbents will not
only provide a clean gas, but also provide pure NG for use. Higher
removal efficiencies were obtained for H2S than for CO2. 10–20%
amine-alkaline mix gave poor results because, they are lower than
the optimum concentration required for the operation with pres-
sure above the critical value of 5250 kPa.
Declaration of Competing Interest
The authors declare that they have no known competing finan-
cial interests or personal relationships that could have appeared
to influence the work reported in this paper.
Acknowledgement
The authors wish to appreciate Exploration and Production
Nigeria Limited for their assistance during the field/bench work
of this research.
References
[1] A. Demirbas, Natural Gas: Methane Gas Hydrate. Springer, Ch. 2, pp. 57-76,
2010. SBN: 978-1-84882-871-1. http://www.springer.com
[2] International Energy Agency, World Energy Outlook, International Energy
Agency, 2010. http://www.worldenergyoutlook.org.
[3] D. Huo, The Global Sour Gas Problem, Energy Resource Engineering, Stanford
University, 2012.
[4] O. Bolland, CO2 Capture in Power Plants, Norwegian University of Science and
Technology, 2013. http://www.ivt.ntnu.no/ept/fag/fordypn/tep03/innhold/
EP03_Part4_Absorption.pdf.
[5] C. Carpenter, J. Petrol. Technol. 67 (2015) 1–11.
[6] A. Erfani, S. Boroojerdi, A. Dehghani, Petrol. Coal 57 (1) (2015) 85–92.
[7] M. Fang, D. Zhu, Chemical Absorption. In: Handbook of Climate Change
Mitigation, Springer, New York Dordrecht Heidelberg London, 2012. ISBN 978-
1-4419-7990-2.
[8] M. Harasimowicz, P. Orluk, G. Zakrzewska-Trznadel, A. Chmielewski, J.
Hazardous Mater. 144 (3) (2007) 698–702.
[9] U. Marchaim, Biogas processes for sustainable development, FAO, Rome, Italy,
1992.
[10] L. Hien, International Conference on Effects of Gas Composition – Aug 02, 2002,
CFS International Pty Ltd.
[11] V.K. Vijay, R. Chandra, P.M.V. Subbarao, S. Kapid, 2nd Joint International
Conference on Sustainable Energy and Environment (SEE), 2006, Bangkok,
Thailand.
[12] D.P. Gosh, The Tri-Service Corrosion Conference, December 3-7, 2007, Denver,
Colorado, USA.
Please cite this article as: S. Eshorame Sanni, O. Agboola, O. Fagbiele et al., Optimization of natural gas treatment for the removal of CO2 and H2S in a novel
alkaline-DEA hybrid scrubber, Egyptian Journal of Petroleum, https://doi.org/10.1016/j.ejpe.2019.11.003
[13] China National Corporation, Sour Natural Gas Corrosion Protection and
Purification, Dongzhimen North Street, Beijing, China, 2012.
[14] L.T. Popoola, S.A. Grema, G.K. Latinwo, B. Gutti, A.S, Int. J. Ind. Chem., 4, 2013,
1–15.
[15] J. Lee, Soft Systems. Retrieved March 24, 2017 from Oil and Gas Blog, 2015.
https//croftsystems.net/oil-gas-blog/what-is-acid-gas-sour-gas.
[16] R.E. Treybal, Operaciones de Transferencia de Masa, 2nd ed., Mc Graw Hill,
Mexico, 1996.
[17] J.I. Huertas, N. Giraldo, S. Izquierdo, Removal of H2S and CO2 from biogas by
amine absorption, Intech. Open, China (2011).
[18] National Energy Laboratory, Syngas Contaminant Removal and Conditioning,
Retrieved March 27, 2017. https://www.netldoe.gov/research/coal/energy-
systems/gasification/gasifipedia/agr.
[19] A. Nobel, Functional Chemicals. Alliance Chemicals, USA, 2017. https://
alliancechemicals.com/wpcontent/uploads/2011/09/DEAspecification.pdf.
[20] C. Perry, Activated Carbon Filtration of Amine and Glycol Solutions. Gas
Conditioning Conference, University of Oklahoma Extension Division,
Oklahoma, USA, 1994.
[21] Petropedia Inc., Acid Gas. Petropedia, USA, 2017. http://www.petropedia.com/
definition/88/acid-gas.
[22] R. Rivera-Tinoco, C. Bouallou, J. Clean. Product. 18 (2010) 875–880.
[23] N. Rodriguez, S. Mustati, N. Scenna, Chem. Eng. Res. Des. 89 (2011) (2011)
1763–1773.
[24] A.E. Salako, Removal of Carbon Dioxide from Natural Gas for LNG Production,
Trondheim, Norwegian University of Science and TechnologyInstitute of
Petroleum Technology, Norway, 2005, https://pdfs.semanticscholar.org/37ea/
1e656cc310e4797ded37b8781dd06d97694c.pdf.
[25] A. Kohl, R. Nielsen, Gas Purification, 5th., Gulf Publishing Company, Houston
Texas, 1997.
[26] P.H. Santana, S.V. Bajay, Chem. Eng. Res. Des. 2 (2016) 22–26.
[27] R. Tatin, L. Moura, N. Dietrich, S. Baig, G. Hebrard 104 (2015) 409–415.
[28] M.S. Horikawa, M.L. Rossi, M.L. Gimenes, C.M.M. Costa, M.G.C. da Silva, Braz. J.
Chem. Eng. 21 (3) (2004) 415–422.
[29] T. Zucca, L.A. Pellegrini, C. Nava, R. Bellasio, R. Bianconi, Chem, Eng. Trans. 6
(2005) 629–634.
[30] L.A. Pellegrini, S. Gamba, S. Moioli, Ind. Eng. Chem. Res. 49 (2010) 4923–4932.
[31] L.A. Pellegrini, S. Moioli, S. Gamba, Chem. Eng. Res. Design, (2010b).
[32] M.B. Hagg, Chem. Eng. J. 172 (2011) 952–960.
[33] C.A. Scholes, G.W. Stevens, S.E. Kentish, Fuel 96 (2012) 15–28.
[34] J.F. Palomeque-santiago, J. Guzman, A.J. Zuniga-mendiola, Revista Mexicana De
Ingeniería Química 15 (2) (2016) 611–624.
[35] B. Belaissaoui, E. Favre, Separat. Purificat. Technol. 193 (2018) 112–126.
[36] Aspen Tech, Modelling and Simulation using HYSYS: Training Manual, Aspen
Tech, Australia, 2006.
[37] AspenTech, Physical Property System: Physical Property Models, Version 7,
Aspen Tech. Inc., 2010.
[38] R.L. Kent, B. Eisenberg, Hydro. Proc. 55 (2) (1976) 87–90.
[39] R. Tennyson, R. Schaff, Oil Gas J. (1997) 78–85.
[40] J.D. Seader, E.J. Henley, Separation Process Principles, 2nd Edition., John Wiley
and Sons, USA, 2006.
[41] D. Peng, D.B. Robinson, Ind. Eng. Chem. Fund. 15 (1976) 59–63.
[42] J. Hao, P.A. Rice, S.A. Stern, J. Membrane Sci. 320 (2008) 108–122.
[43] E.S. Van-Dal, C. Bouallou, J. Cleaner Production 57 (2013) 38–45.
[44] H. Lababidi, G.A. Al-Enezi, H.M. Ettouney, J. Membrane Sci. 112 (1996) 185–
197.
[45] M. Al-Marzouqi, S. Marzouk, N. Abdullatif, Removal of Acid Gases from Natural
Gas using Gas-Liquid Contactors, Microsoft Power point -1-2 GCC-JCCP-Al-
Mazouqi, UAE University, 2014.
[46] B. Pouladi, M.N. Hassankiadeh, F. Behroozshada, Energy Reports 2 (2016) 54–
61.
[47] Lunsford, K.; Bullin, J.A. Optimization of Amine Sweetening Units, Proceedings
of the AIChE Spring National Meeting, New York, USA, 1996.
[48] Caothien, S., Hasemit, R.; Pauley, C.R. Analysis of Foaming Mechanisms in
Amine Plants, Proceedings of the AIChE Summer Meeting, Denver, Colorado,
Canada, 1988.
[49] Bourke, M.J.; Mazzoni, A.F. ‘‘The Roles of Activated Carbon in Gas
Conditioning”, Proceedings of the Laurence Reid Gas Conditioning
Conference, Norman, Oklahoma, 1989.
[50] Addington, L.; Ness, C. An Evaluation of General ‘‘Rules of Thumb” in Amine
Sweetening Unit Design and Operation. Bryan Research and Engineering, Inc.
Bryan, Texas, USA, 2010.
[51] K. Lunsford, G. Mcintyre, Decreasing Contactor Temperature Could Increase
Performance, Bryan Research and Engineering, Inc. – Technical Papers, Bryan
Texas, USA, 2006.
[52] F.B.A. Basri, Bachelor of Chemical Engineering Thesis, Faculty of Chemical &
Natural Resources Engineering Universiti Malaysia Pahang, Malaysia, 2014.
[53] Z.H. Aliabad, S. Mirzaei, Int. J. Chem. Molecular Eng. 3 (1) (2009) 50–59.
[54] S.A. Rackley, Carbon Capture and Storage. Elsevier Inc., 2017. https://doi.org/
10.1016/C2015-0-01587-8. ISBN: 9780128120415.
[55] S.-J. Han, M. Yoo, D.-W. Kim, J.-H. Wee, Energy Fuels 25 (2011) 3825–3834.
[56] M. Chandel, E. Williams, Synthetic Natural Gas (SNG): Technology,
Environmental Implications, and Economics, Duke University, Climate
Change Policy Partnership, 2009.